Data Quality Models for High Volume Transaction Streams: A Case Study Robert Grossman, Chris Curry, David Locke & Steve Vejcik Open Data Group Joseph Bugajski.

Slides:



Advertisements
Similar presentations
Credit Card Processing 101
Advertisements

Evolving Challenges of PCI Compliance Charlie Wood, PCI QSA, CRISC, CISA Principal, The Bonadio Group January 10, 2014.
Purchasing Assets with a P-CARD Presented by: David Bonola Tom Vaughan.
Tools for Transforming Ideas into Strategy By Eliza Chute and Kym Cole.
Credit Card Fraud The Scale of the Problem Michael Moore Regional Security & Fraud Investigation Manager 14 – 17 Nov 2005 Security & Safety – Middle East.
IAB Affiliate Advertiser Survey In association with A4U October 2012.
Vice President, e-Business Development Dubai United Nations Conference on Trade & Development Conference on Electronic Commerce.
All Roads Lead to China. Travel Payments Direct  Founded in February 2013 by a core group of payments professionals possessing 20-years of collective.
International Card Systems Skopje, Macedonia
How to do Business Online - Securely Presented by: Michael Gulliver First Data Regional Credit and Risk Director.
CSU Kuali LabTraining Welcome to Kuali Lab Training! Lab Training Session for: Kuali PCARD June 2009.
Joe SimonettiT-FLEx Workshop T-FLEx October Workshop The Future of Fare Collection Bank Card Transactions & Merchant Processing Joseph Simonetti October.
Discover Financial NYSE: DFS Phil Rosen Tim Tyson Xiaoting Yang Presented,
ZETA ERP FINANCE MODULE. Finance General Ledger Accounts receivable Accounts Payable Bank FINANCE MODULE.
PCI's Changing Environment – “What You Need to Know & Why You Need To Know It.” Stephen Scott – PCI QSA, CISA, CISSP
Cards and ePayments Brian Hartzer Managing Director Australia and New Zealand Banking Group Limited 24 August 2001.
© Sam Ransbotham The Impact of Immediate Disclosure on Attack Diffusion and Volume Sam Ransbotham Boston College Sabyasachi Mitra Georgia Institute of.
Business Model Canvas From: Business Model Generation Preview
People © 2013 The Sleeter Group All rights reserved. Intuit, the Intuit logo and QuickBooks, among others, are registered trademarks of Intuit Inc. Other.
2013.  Reconcile your checking account  Create bank reconciliation reports  Find errors during reconciliation  Correct errors found during reconciliation.
Corporate Purchasing Card Enhanced Reporting January 2015 Web Version 1.
Data Mining Techniques
CS490D: Introduction to Data Mining Prof. Chris Clifton April 14, 2004 Fraud and Misuse Detection.
Achieving Better Reliability With Software Reliability Engineering Russel D’Souza Russel D’Souza.
PKI Issues: The Payment Perspective March 6, 2000 Ann Terwilliger eCommerce Authentication Visa International.
LU Chenglong ( ) DIAO Wenrui ( )
Federal Acquisition Service U.S. General Services Administration Bradley Forrestel Elizabeth Skolnik Contract Specialists General Services Administration.
Linking the World Through Learning 1 GEM – GDLN Event Management system GDLN Asia Pacific General Meeting, June 2007.
Smart Payment Processing ™ Recur} Happen again. Persist. Return. Come back. Reappear. Come again.
GRAPHITE PAYMENTS ATM PROGRAM 1. ATM STATISTICS 2013 Federal Reserve Payments Study  5.8 billion ATM withdrawals  1.9 billion from non-FI ATMs  $
S T R I C T L Y P R I V A T E A N D C O N F I D E N T I A LS T R I C T L Y P R I V A T E A N D C O N F I D E N T I A L 0 S T R I C T L Y P R I V A T E.
THE MOBILE CHANNEL IN FINANCIAL SERVICES TARIK HUSAIN BUSINESS DEVELOPMENT DIRECTOR ASIAN BANKER SUMMIT APRIL 2011.
2007 ACM SIGKDD Data Mining Practice Prize Winners Brendan Kitts (co-chair), iProspect Gabor Melli (co-chair), Simon Fraser University Gregory Piatetsky-Shapiro,
Integration Points Basic Setup Payroll Processing Key Implementation Tips Session focus.
PCard Training Logging into the new PCard (PaymentNet) System: www5.paymentnet.com.
2014 Asia-Pacific Financial Forum Seattle, Washington July 7, 2014 Electronic Payments: Expanding Financial Access for Consumers and Businesses of Every.
Business Models.
A statistical test for point source searches - Aart Heijboer - AWG - Cern june 2002 A statistical test for point source searches Aart Heijboer contents:
Next Generation of Online Banking and Bill Pay. 2 © 2010 – Proprietary & Confidential The Next Generation of Online Banking and Bill Pay is Here!
Confidential © 2014, Brighterion Inc. (all rights reserved) Keeping an eye on your business Brighterion Solutions SMART AGENTS SMART AGENTS UNSUPERVISED.
GNS Science Natural Hazards Research Platform Progress in understanding the Canterbury Earthquakes Kelvin Berryman Manager, Natural Hazards Research Platform.
Secure In-Network Aggregation for Wireless Sensor Networks
Real-time Intelligence that Matters 1. © 2015, Brighterion Inc. (all rights reserved) Keeping an eye on your business 53% of consumers who experienced.
SABRE VIRTUAL PAYMENTS Karen Frayer Sabre Virtual Payments Manager.
More Than Relevance: High Utility Query Recommendation By Mining Users' Search Behaviors Xiaofei Zhu, Jiafeng Guo, Xueqi Cheng, Yanyan Lan Institute of.
VeriFone Virtual Terminal Web-Hosted Hosted Payment Gateway
Detection, Classification and Tracking in Distributed Sensor Networks D. Li, K. Wong, Y. Hu and A. M. Sayeed Dept. of Electrical & Computer Engineering.
MudiamPCI provide the solution for SAP credit card processing, payment card and card tokenization with aes 256 encryption.
A very important business in the EU… 508 million People 760 million Cards 43.6 billion Transactions 2.2 trillion Euros Source: European Central Bank Press.
Salient features of facility:  Minimum amount of withdrawal Rs.100/- (thereafter in multiples of Rs.100/- ).  Maximum of Rs.1000/- per day per.
The target size of the fund and the methodology of estimation of DI Fund sufficiency July, 2015.
TRACE ANALYSIS AND MINING FOR SMART CITIES By G. Pan Zhejiang Univ., Hangzhou, China G. Qi ; W. Zhang ; S. Li ; Z. Wu ; L. T. Yang.
Who’s A Control Freak? Audrey Flood November 2006 State of Texas User Conference.
Page 1 of 17 M. Ufuk Caglayan, CmpE 476 Spring 2000, SSL and SET Notes, March 29, 2000 CmpE 476 Spring 2000 Notes on SSL and SET Dr. M. Ufuk Caglayan Department.
Created by BM|DESIGN|ER e-Bookkeeping default Home Corporate Stage 2.
Big Data, Commercial Data, and the National Accounts Brent Moulton Advisory Expert Group on National Accounts 13–15 April 2016 Note: Slides are drawn from.
 Accounting Improvements What’s New with CiviCRM and Accounting Integration Joe Murray, President Presented at CiviCon Colorado, June 2016.
New Media – Keys to Surfing
Customer Analytics: Strategies for Success
Credit Card Processing 101
Presented by: John Flynn, First Data Heather Fletcher, First Data
Mastercard Location Alerts™
Checking Services and Credit-Card Transactions
Checking Services and Credit-Card Transactions
Merchant Payments ITU Digital Financial Services Workshop
Checking Services and Credit- Card Transactions
Online Payment Options for Government
Presentation transcript:

Data Quality Models for High Volume Transaction Streams: A Case Study Robert Grossman, Chris Curry, David Locke & Steve Vejcik Open Data Group Joseph Bugajski Visa International

The Problem: Detect Significant Changes in Visa’s Payments Network Account Merchant Issuing Bank Acquiring Bank

Visa Payment Network  Over 1.59 billion Visa cards in circulation  6800 transactions per second (peak)  Over 20,000 member banks  Millions of merchants

The Challenge: Payments Data is Highly Heterogeneous  Variation from cardholder to cardholder  Variation from merchant to merchant  Variation from bank to bank

Observe: If Data Were Homogeneous, Could Use Change Detection Model  Sequence of events x[1], x[2], x[3], …  Question: is the observed distribution different than the baseline distribution?  Use simple CUSUM & Generalized Likelihood Ratio (GLR) tests Observed Model Baseline Model 

Key Idea: Build Models, One for Each Cell in Data Cube  Build separate model for each bank (1000+)  Build separate model for each geographical region (6 regions)  Build separate model for each different type of merchant (c. 800 types of merchants)  For each distinct cube, establish separate baselines for each metric of interest (declines, etc.)  Detect changes from baselines Geospatial region Type of Transaction 20,000+ separate baselines Modeling using Cubes of Models (MCM) Bank

Greedy Meaningful/Manageable Balancing (GMMB) Algorithm Fewer alerts Alerts more manageable To decrease alerts, remove breakpoint, order by number of decreased alerts, & select one or more breakpoints to remove More alerts Alerts more meaningful To increase alerts, add breakpoint to split cubes, order by number of new alerts, & select one or more new breakpoints One model for each cell in data cube Breakpoint

Augustus  Open source Augustus data mining platform was used to: Estimate baselines for over 15,000 separate segmented models Score high volume operational data and issue alerts for follow up investigations  Augustus is PMML compliant  Augustus scales with Volume of data (Terabytes) Real time transaction streams (15,000/sec+) Number of segmented models (10,000+)

Some Results to Date  System has been operational for 2.5 years  ROI –5.1xYear 1(over 6 months) –7.3xYear 2(12 months) –10.0x Year 3(12 months)  Currently estimating over 15,000 individual baseline models  The system has issued alerts for: Merchants using incorrect Merchant Category Code (MCC) - account testing Sales channel variations Incorrect use of merchant city name field Incorrect coding or recurring payments

Summary  Used new methodology (Modeling using Cubes of Models) for modeling large, highly heterogeneous data sets  This project contributed in part to the development of a Baseline Model in the open source Augustus system  Integrated system to generate alerts using Baseline Models with manual investigation process  Project is generating over 10x ROI  Poster #20 in the Tuesday night Poster Session.